2.0 Analysis 2.1 Introduction The investigation revealed that both the pilot and co-pilot were properly licensed and qualified, and the aircraft was serviceable for the flight. 2.2 The Power Loss In consideration of the recent maintenance history of the aircraft, the engines were inspected and test runs were conducted to determine what could have caused a total or partial power loss on one or both of the engines. The engine test runs indicated that both engines were capable of producing full power even when tested in an almost as recovered condition. Based on the inspection of the engines and their performance during the test runs, the cause of the total or partial power loss on one or both of the engines could not be duplicated or determined. The examination of the left propeller engine flange, the right propeller, and available cockpit engine instruments indicated that neither propeller was feathered at impact and that the left engine was producing at least partial power. The impact mark on the right engine tachometer indicated that the right propeller was at 2,450 rpm at impact. It is most likely that the power loss experienced by the pilot was not caused by a mechanical malfunction of the engines. 2.3 Aircraft Performance Although the nature and source of the power loss experienced by the pilot could not be determined, the examination of the radar data did reveal several key aspects about the aircraft's performance. The radar data indicated that the initial climb was normal to 1,050 feet asl, at which time the ground speed decreased. This could have occurred when the pilot was reducing to climb power. Since the aircraft's rate of climb was a constant 1,000 fpm, the airspeed would have decreased as a result of the lower power setting. The aircraft momentarily levelled off at 1,200 feet asl. When the pilot made his first call to the departure controller through about 1,300 feet asl, he did not indicate that he was having any engine problems. Therefore, it is unlikely that the momentary level-off at 1,200 feet asl was a result of the engine failure or the loss of power. Since the pilot had difficulty with IFR flight, he may have found it advantageous to use the autopilot at this busy time of the flight. The momentary level-off could have been caused by the pilot engaging the autopilot if the autopilot was set for level flight. If the pilot then activated the autopilot trim wheel for a climb, this could explain the climb from 1,200 feet asl to 1,500 feet asl with a constant rate of climb of 1,000 fpm while the airspeed remained low. If the autopilot was engaged, the pilot would have been free to call the departure controller, complete his after take-off checks, and adjust the power settings. If the pilot or co-pilot leaned the mixtures using the newly installed Shadin Digiflo-L fuel management system during this part of the climb, a power loss situation may have inadvertently occurred. If the mixtures were reduced to a fuel flow setting appropriate to the smaller 285 hp engines of the aircraft that the pilot owned and was familiar with, then, given the greater fuel flow required by the 310 hp engines on the accident aircraft, it is possible that the accident aircraft could have lost partial or total power on one or both engines. If this occurred, and the aircraft autopilot was engaged, the pilot may have been distracted in dealing with the engine malfunction and not have noticed the airspeed decrease. The radar data indicated that, at 1,500 feet asl, the aircraft descended rapidly with no increase in groundspeed; this could have resulted from the overweight aircraft stalling. The aircraft descended to 600 feet asl before the rate of descent was arrested, then a 1,000 fpm climb was briefly re-established as the aircraft climbed to an altitude of 900 feet asl and the ground speed decreased to 83 knots. Passing through 800 feet asl, the pilot informed the departure controller that he had an engine failure and wanted to immediately return to the airport. This was the last communication from the aircraft. The last radar target showed the aircraft was at 900 feet asl in a right turn. The overweight aircraft most likely stalled again, and the pilot had insufficient altitude to recover as it descended steeply out of control into Lake Ontario. 2.4 Weather Factors The lowest cloud near the airport at the time of the occurrence was a scattered layer at 1,450 feet asl. The next layer of cloud was at 1,750 feet asl. Since the aircraft climbed to 1,500 feet asl, it is possible that the aircraft entered cloud. The visibility at the time was as low as four miles in fog. These weather conditions could have aggravated the pilot's ability to regain control of the aircraft; the lack of a discernible horizon would result in disorientation, particularly since the pilot had weak instrument flying skills. 3.0 Conclusions 3.1 Findings The aircraft was 345.3 pounds above the maximum gross take-off weight when the flight departed, and the aircraft was operating outside of the approved weight and balance envelope at the time of the accident. Both the pilot and co-pilot were properly licensed and qualified to fly the aircraft. The aircraft was maintained in accordance with approved procedures and regulations. The aircraft experienced a power loss during the initial climb-out. The extent and nature of the power loss was not determined; however, the power loss may have been induced by one of the pilots. The pilot lost control of the overweight aircraft at 1,500 feet asl, while operating in cloud, and descended to 600 feet asl prior to regaining control of the aircraft. This was followed by a second loss of control at 900 feet asl. Since the pilot had weak instrument flying skills, the weather conditions at the time of the occurrence may have aggravated the pilot's ability to recover the aircraft. The aircraft struck the water in a steep, nose-down, left-wing-low attitude. 3.2 Causes After experiencing a power loss during the initial climb-out, the pilot lost control of the overweight aircraft while attempting to return to the airport. The cause of the power loss was not determined; however, both engines were found to be capable of producing full power when tested. The Board has no aviation safety recommendations to issue at this time.4.0 Safety Action The Board has no aviation safety recommendations to issue at this time.